3 Fix memory leak in reconnect. Fix oops in DFS mount error path.
4 Set s_maxbytes to smaller (the max that vfs can handle) so that
5 sendfile will now work over cifs mounts again. Add noforcegid
6 and noforceuid mount parameters.
10 Client uses server inode numbers (which are persistent) rather than
11 client generated ones by default (mount option "serverino" turned
12 on by default if server supports it). Add forceuid and forcegid
13 mount options (so that when negotiating unix extensions specifying
14 which uid mounted does not immediately force the server's reported
15 uids to be overridden). Add support for scope mount parm. Improve
16 hard link detection to use same inode for both. Do not set
17 read-only dos attribute on directories (for chmod) since Windows
18 explorer special cases this attribute bit for directories for
23 Guard against buffer overruns in various UCS-2 to UTF-8 string conversions
24 when the UTF-8 string is composed of unusually long (more than 4 byte) converted
25 characters. Add support for mounting root of a share which redirects immediately
26 to DFS target. Convert string conversion functions from Unicode to more
27 accurately mark string length before allocating memory (which may help the
28 rare cases where a UTF-8 string is much larger than the UCS2 string that
29 we converted from). Fix endianness of the vcnum field used during
30 session setup to distinguish multiple mounts to same server from different
31 userids. Raw NTLMSSP fixed (it requires /proc/fs/cifs/experimental
32 flag to be set to 2, and mount must enable krb5 to turn on extended security).
33 Performance of file create to Samba improved (posix create on lookup
34 removes 1 of 2 network requests sent on file create)
38 Improve support for multiple security contexts to the same server. We
39 used to use the same "vcnumber" for all connections which could cause
40 the server to treat subsequent connections, especially those that
41 are authenticated as guest, as reconnections, invalidating the earlier
42 user's smb session. This fix allows cifs to mount multiple times to the
43 same server with different userids without risking invalidating earlier
44 established security contexts. fsync now sends SMB Flush operation
45 to better ensure that we wait for server to write all of the data to
46 server disk (not just write it over the network). Add new mount
47 parameter to allow user to disable sending the (slow) SMB flush on
48 fsync if desired (fsync still flushes all cached write data to the server).
49 Posix file open support added (turned off after one attempt if server
50 fails to support it properly, as with Samba server versions prior to 3.3.2)
51 Fix "redzone overwritten" bug in cifs_put_tcon (CIFSTcon may allocate too
52 little memory for the "nativeFileSystem" field returned by the server
53 during mount). Endian convert inode numbers if necessary (makes it easier
54 to compare inode numbers on network files from big endian systems).
58 Add "forcemandatorylock" mount option to allow user to use mandatory
59 rather than posix (advisory) byte range locks, even though server would
60 support posix byte range locks. Fix query of root inode when prefixpath
61 specified and user does not have access to query information about the
62 top of the share. Fix problem in 2.6.28 resolving DFS paths to
63 Samba servers (worked to Windows). Fix rmdir so that pending search
64 (readdir) requests do not get invalid results which include the now
65 removed directory. Fix oops in cifs_dfs_ref.c when prefixpath is not reachable
66 when using DFS. Add better file create support to servers which support
67 the CIFS POSIX protocol extensions (this adds support for new flags
68 on create, and improves semantics for write of locked ranges).
72 Various fixes to make delete of open files behavior more predictable
73 (when delete of an open file fails we mark the file as "delete-on-close"
74 in a way that more servers accept, but only if we can first rename the
75 file to a temporary name). Add experimental support for more safely
76 handling fcntl(F_SETLEASE). Convert cifs to using blocking tcp
77 sends, and also let tcp autotune the socket send and receive buffers.
78 This reduces the number of EAGAIN errors returned by TCP/IP in
79 high stress workloads (and the number of retries on socket writes
80 when sending large SMBWriteX requests). Fix case in which a portion of
81 data can in some cases not get written to the file on the server before the
82 file is closed. Fix DFS parsing to properly handle path consumed field,
83 and to handle certain codepage conversions better. Fix mount and
84 umount race that can cause oops in mount or umount or reconnect.
88 Fix premature write failure on congested networks (we would give up
89 on EAGAIN from the socket too quickly on large writes).
90 Cifs_mkdir and cifs_create now respect the setgid bit on parent dir.
91 Fix endian problems in acl (mode from/to cifs acl) on bigendian
92 architectures. Fix problems with preserving timestamps on copying open
93 files (e.g. "cp -a") to Windows servers. For mkdir and create honor setgid bit
94 on parent directory when server supports Unix Extensions but not POSIX
95 create. Update cifs.upcall version to handle new Kerberos sec flags
96 (this requires update of cifs.upcall program from Samba). Fix memory leak
97 on dns_upcall (resolving DFS referralls). Fix plain text password
98 authentication (requires setting SecurityFlags to 0x30030 to enable
99 lanman and plain text though). Fix writes to be at correct offset when
100 file is open with O_APPEND and file is on a directio (forcediretio) mount.
101 Fix bug in rewinding readdir directory searches. Add nodfs mount option.
105 DFS support added (Microsoft Distributed File System client support needed
106 for referrals which enable a hierarchical name space among servers).
107 Disable temporary caching of mode bits to servers which do not support
108 storing of mode (e.g. Windows servers, when client mounts without cifsacl
109 mount option) and add new "dynperm" mount option to enable temporary caching
110 of mode (enable old behavior). Fix hang on mount caused when server crashes
111 tcp session during negotiate protocol.
115 Fix oops on second mount to server when null auth is used.
116 Enable experimental Kerberos support. Return writebehind errors on flush
117 and sync so that events like out of disk space get reported properly on
118 cached files. Fix setxattr failure to certain Samba versions. Fix mount
119 of second share to disconnected server session (autoreconnect on this).
120 Add ability to modify cifs acls for handling chmod (when mounted with
121 cifsacl flag). Fix prefixpath path separator so we can handle mounts
122 with prefixpaths longer than one directory (one path component) when
123 mounted to Windows servers. Fix slow file open when cifsacl
124 enabled. Fix memory leak in FindNext when the SMB call returns -EBADF.
129 Fix memory leak in statfs when mounted to very old servers (e.g.
130 Windows 9x). Add new feature "POSIX open" which allows servers
131 which support the current POSIX Extensions to provide better semantics
132 (e.g. delete for open files opened with posix open). Take into
133 account umask on posix mkdir not just older style mkdir. Add
134 ability to mount to IPC$ share (which allows CIFS named pipes to be
135 opened, read and written as if they were files). When 1st tree
136 connect fails (e.g. due to signing negotiation failure) fix
137 leak that causes cifsd not to stop and rmmod to fail to cleanup
138 cifs_request_buffers pool. Fix problem with POSIX Open/Mkdir on
139 bigendian architectures. Fix possible memory corruption when
140 EAGAIN returned on kern_recvmsg. Return better error if server
141 requires packet signing but client has disabled it. When mounted
142 with cifsacl mount option - mode bits are approximated based
143 on the contents of the ACL of the file or directory. When cifs
144 mount helper is missing convert make sure that UNC name
145 has backslash (not forward slash) between ip address of server
150 Fix NTLMv2 signing. NFS server mounted over cifs works (if cifs mount is
151 done with "serverino" mount option). Add support for POSIX Unlink
152 (helps with certain sharing violation cases when server such as
153 Samba supports newer POSIX CIFS Protocol Extensions). Add "nounix"
154 mount option to allow disabling the CIFS Unix Extensions for just
155 that mount. Fix hang on spinlock in find_writable_file (race when
156 reopening file after session crash). Byte range unlock request to
157 windows server could unlock more bytes (on server copy of file)
158 than intended if start of unlock request is well before start of
159 a previous byte range lock that we issued.
163 IPv6 support. Enable ipv6 addresses to be passed on mount (put the ipv6
164 address after the "ip=" mount option, at least until mount.cifs is fixed to
165 handle DNS host to ipv6 name translation). Accept override of uid or gid
166 on mount even when Unix Extensions are negotiated (it used to be ignored
167 when Unix Extensions were ignored). This allows users to override the
168 default uid and gid for files when they are certain that the uids or
169 gids on the server do not match those of the client. Make "sec=none"
170 mount override username (so that null user connection is attempted)
171 to match what documentation said. Support for very large reads, over 127K,
172 available to some newer servers (such as Samba 3.0.26 and later but
173 note that it also requires setting CIFSMaxBufSize at module install
174 time to a larger value which may hurt performance in some cases).
175 Make sign option force signing (or fail if server does not support it).
179 Fix mtime bouncing around from local idea of last write times to remote time.
180 Fix hang (in i_size_read) when simultaneous size update of same remote file
181 on smp system corrupts sequence number. Do not reread unnecessarily partial page
182 (which we are about to overwrite anyway) when writing out file opened rw.
183 When DOS attribute of file on non-Unix server's file changes on the server side
184 from read-only back to read-write, reflect this change in default file mode
185 (we had been leaving a file's mode read-only until the inode were reloaded).
186 Allow setting of attribute back to ATTR_NORMAL (removing readonly dos attribute
187 when archive dos attribute not set and we are changing mode back to writeable
188 on server which does not support the Unix Extensions). Remove read only dos
189 attribute on chmod when adding any write permission (ie on any of
190 user/group/other (not all of user/group/other ie 0222) when
191 mounted to windows. Add support for POSIX MkDir (slight performance
192 enhancement and eliminates the network race between the mkdir and set
193 path info of the mode).
198 Fix oops in list_del during mount caused by unaligned string.
199 Fix file corruption which could occur on some large file
200 copies caused by writepages page i/o completion bug.
201 Seek to SEEK_END forces check for update of file size for non-cached
202 files. Allow file size to be updated on remote extend of locally open,
203 non-cached file. Fix reconnect to newer Samba servers (or other servers
204 which support the CIFS Unix/POSIX extensions) so that we again tell the
205 server the Unix/POSIX cifs capabilities which we support (SetFSInfo).
206 Add experimental support for new POSIX Open/Mkdir (which returns
207 stat information on the open, and allows setting the mode).
211 Support deep tree mounts. Better support OS/2, Win9x (DOS) time stamps.
212 Allow null user to be specified on mount ("username="). Do not return
213 EINVAL on readdir when filldir fails due to overwritten blocksize
214 (fixes FC problem). Return error in rename 2nd attempt retry (ie report
215 if rename by handle also fails, after rename by path fails, we were
216 not reporting whether the retry worked or not). Fix NTLMv2 to
217 work to Windows servers (mount with option "sec=ntlmv2").
221 Do not time out lockw calls when using posix extensions. Do not
222 time out requests if server still responding reasonably fast
223 on requests on other threads. Improve POSIX locking emulation,
224 (lock cancel now works, and unlock of merged range works even
225 to Windows servers now). Fix oops on mount to lanman servers
226 (win9x, os/2 etc.) when null password. Do not send listxattr
227 (SMB to query all EAs) if nouser_xattr specified. Fix SE Linux
228 problem (instantiate inodes/dentries in right order for readdir).
232 Rewritten sessionsetup support, including support for legacy SMB
233 session setup needed for OS/2 and older servers such as Windows 95 and 98.
234 Fix oops on ls to OS/2 servers. Add support for level 1 FindFirst
235 so we can do search (ls etc.) to OS/2. Do not send NTCreateX
236 or recent levels of FindFirst unless server says it supports NT SMBs
237 (instead use legacy equivalents from LANMAN dialect). Fix to allow
238 NTLMv2 authentication support (now can use stronger password hashing
239 on mount if corresponding /proc/fs/cifs/SecurityFlags is set (0x4004).
240 Allow override of global cifs security flags on mount via "sec=" option(s).
244 POSIX locking to servers which support CIFS POSIX Extensions
245 (disabled by default controlled by proc/fs/cifs/Experimental).
246 Handle conversion of long share names (especially Asian languages)
247 to Unicode during mount. Fix memory leak in sess struct on reconnect.
248 Fix rare oops after acpi suspend. Fix O_TRUNC opens to overwrite on
249 cifs open which helps rare case when setpathinfo fails or server does
254 Fix slow oplock break when mounted to different servers at the same time and
255 the tids match and we try to find matching fid on wrong server. Fix read
256 looping when signing required by server (2.6.16 kernel only). Fix readdir
257 vs. rename race which could cause each to hang. Return . and .. even
258 if server does not. Allow searches to skip first three entries and
259 begin at any location. Fix oops in find_writeable_file.
263 Fix NTLMv2 security (can be enabled in /proc/fs/cifs) so customers can
264 configure stronger authentication. Fix sfu symlinks so they can
265 be followed (not just recognized). Fix wraparound of bcc on
266 read responses when buffer size over 64K and also fix wrap of
267 max smb buffer size when CIFSMaxBufSize over 64K. Fix oops in
268 cifs_user_read and cifs_readpages (when EAGAIN on send of smb
269 on socket is returned over and over). Add POSIX (advisory) byte range
270 locking support (requires server with newest CIFS UNIX Extensions
271 to the protocol implemented). Slow down negprot slightly in port 139
272 RFC1001 case to give session_init time on buggy servers.
276 Use fsuid (fsgid) more consistently instead of uid (gid). Improve performance
277 of readpages by eliminating one extra memcpy. Allow update of file size
278 from remote server even if file is open for write as long as mount is
279 directio. Recognize share mode security and send NTLM encrypted password
280 on tree connect if share mode negotiated.
284 Defer close of a file handle slightly if pending writes depend on that handle
285 (this reduces the EBADF bad file handle errors that can be logged under heavy
286 stress on writes). Modify cifs Kconfig options to expose CONFIG_CIFS_STATS2
287 Fix SFU style symlinks and mknod needed for servers which do not support the
288 CIFS Unix Extensions. Fix setfacl/getfacl on bigendian. Timeout negative
289 dentries so files that the client sees as deleted but that later get created
290 on the server will be recognized. Add client side permission check on setattr.
291 Timeout stuck requests better (where server has never responded or sent corrupt
296 Fix tcp socket retransmission timeouts (e.g. on ENOSPACE from the socket)
297 to be smaller at first (but increasing) so large write performance performance
298 over GigE is better. Do not hang thread on illegal byte range lock response
299 from Windows (Windows can send an RFC1001 size which does not match smb size) by
300 allowing an SMBs TCP length to be up to a few bytes longer than it should be.
301 wsize and rsize can now be larger than negotiated buffer size if server
302 supports large readx/writex, even when directio mount flag not specified.
303 Write size will in many cases now be 16K instead of 4K which greatly helps
304 file copy performance on lightly loaded networks. Fix oops in dnotify
305 when experimental config flag enabled. Make cifsFYI more granular.
309 Fix readdir caching when unlink removes file in current search buffer,
310 and this is followed by a rewind search to just before the deleted entry.
311 Do not attempt to set ctime unless atime and/or mtime change requested
312 (most servers throw it away anyway). Fix length check of received smbs
313 to be more accurate. Fix big endian problem with mapchars mount option,
314 and with a field returned by statfs.
318 Add support for mounting to older pre-CIFS servers such as Windows9x and ME.
319 For these older servers, add option for passing netbios name of server in
320 on mount (servernetbiosname). Add suspend support for power management, to
321 avoid cifsd thread preventing software suspend from working.
322 Add mount option for disabling the default behavior of sending byte range lock
323 requests to the server (necessary for certain applications which break with
324 mandatory lock behavior such as Evolution), and also mount option for
325 requesting case insensitive matching for path based requests (requesting
326 case sensitive is the default).
330 Add writepage performance improvements. Fix path name conversions
331 for long filenames on mounts which were done with "mapchars" mount option
332 specified. Ensure multiplex ids do not collide. Fix case in which
333 rmmod can oops if done soon after last unmount. Fix truncated
334 search (readdir) output when resume filename was a long filename.
335 Fix filename conversion when mapchars mount option was specified and
336 filename was a long filename.
340 Fix error mapping of the TOO_MANY_LINKS (hardlinks) case.
341 Do not oops if root user kills cifs oplock kernel thread or
342 kills the cifsd thread (NB: killing the cifs kernel threads is not
343 recommended, unmount and rmmod cifs will kill them when they are
344 no longer needed). Fix readdir to ASCII servers (ie older servers
345 which do not support Unicode) and also require asterisk.
346 Fix out of memory case in which data could be written one page
347 off in the page cache.
351 Fix caching problem, in which readdir of directory containing a file
352 which was cached could cause the file's time stamp to be updated
353 without invalidating the readahead data (so we could get stale
354 file data on the client for that file even as the server copy changed).
355 Cleanup response processing so cifsd can not loop when abnormally
361 Fix oops in ls when Transact2 FindFirst (or FindNext) returns more than one
362 transact response for an SMB request and search entry split across two frames.
363 Add support for lsattr (getting ext2/ext3/reiserfs attr flags from the server)
364 as new protocol extensions. Do not send Get/Set calls for POSIX ACLs
365 unless server explicitly claims to support them in CIFS Unix extensions
366 POSIX ACL capability bit. Fix packet signing when multiuser mounting with
367 different users from the same client to the same server. Fix oops in
368 cifs_close. Add mount option for remapping reserved characters in
369 filenames (also allow recognizing files with created by SFU which have any
370 of these seven reserved characters, except backslash, to be recognized).
371 Fix invalid transact2 message (we were sometimes trying to interpret
372 oplock breaks as SMB responses). Add ioctl for checking that the
373 current uid matches the uid of the mounter (needed by umount.cifs).
374 Reduce the number of large buffer allocations in cifs response processing
375 (significantly reduces memory pressure under heavy stress with multiple
376 processes accessing the same server at the same time).
380 Fix updates of DOS attributes and time fields so that files on NT4 servers
381 do not get marked delete on close. Display sizes of cifs buffer pools in
382 cifs stats. Fix oops in unmount when cifsd thread being killed by
383 shutdown. Add generic readv/writev and aio support. Report inode numbers
384 consistently in readdir and lookup (when serverino mount option is
385 specified use the inode number that the server reports - for both lookup
386 and readdir, otherwise by default the locally generated inode number is used
387 for inodes created in either path since servers are not always able to
388 provide unique inode numbers when exporting multiple volumes from under one
393 Allow new nouser_xattr mount parm to disable xattr support for user namespace.
394 Do not flag user_xattr mount parm in dmesg. Retry failures setting file time
395 (mostly affects NT4 servers) by retry with handle based network operation.
396 Add new POSIX Query FS Info for returning statfs info more accurately.
397 Handle passwords with multiple commas in them.
401 Fix default mode in sysfs of cifs module parms. Remove old readdir routine.
402 Fix capabilities flags for large readx so as to allow reads larger than 64K.
406 Add module init parm for large SMB buffer size (to allow it to be changed
407 from its default of 16K) which is especially useful for large file copy
408 when mounting with the directio mount option. Fix oops after
409 returning from mount when experimental ExtendedSecurity enabled and
410 SpnegoNegotiated returning invalid error. Fix case to retry better when
411 peek returns from 1 to 3 bytes on socket which should have more data.
412 Fixed path based calls (such as cifs lookup) to handle path names
413 longer than 530 (now can handle PATH_MAX). Fix pass through authentication
414 from Samba server to DC (Samba required dummy LM password).
418 Turn off DNOTIFY (directory change notification support) by default
419 (unless built with the experimental flag) to fix hang with KDE
420 file browser. Fix DNOTIFY flag mappings. Fix hang (in wait_event
421 waiting on an SMB response) in SendReceive when session dies but
422 reconnects quickly from another task. Add module init parms for
423 minimum number of large and small network buffers in the buffer pools,
424 and for the maximum number of simultaneous requests.
428 Add setfacl support to allow setting of ACLs remotely to Samba 3.10 and later
429 and other POSIX CIFS compliant servers. Fix error mapping for getfacl
430 to EOPNOTSUPP when server does not support posix acls on the wire. Fix
431 improperly zeroed buffer in CIFS Unix extensions set times call.
435 Fix internationalization problem in cifs readdir with filenames that map to
436 longer UTF-8 strings than the string on the wire was in Unicode. Add workaround
437 for readdir to netapp servers. Fix search rewind (seek into readdir to return
438 non-consecutive entries). Do not do readdir when server negotiates
439 buffer size to small to fit filename. Add support for reading POSIX ACLs from
440 the server (add also acl and noacl mount options).
444 Optionally allow using server side inode numbers, rather than client generated
445 ones by specifying mount option "serverino" - this is required for some apps
446 to work which double check hardlinked files and have persistent inode numbers.
450 Multiple bigendian fixes. On little endian systems (for reconnect after
451 network failure) fix tcp session reconnect code so we do not try first
452 to reconnect on reverse of port 445. Treat reparse points (NTFS junctions)
453 as directories rather than symlinks because we can do follow link on them.
457 Add config option to enable XATTR (extended attribute) support, mapping
458 xattr names in the "user." namespace space to SMB/CIFS EAs. Lots of
459 minor fixes pointed out by the Stanford SWAT checker (mostly missing
460 or out of order NULL pointer checks in little used error paths).
464 Add new mount parm to control whether mode check (generic_permission) is done
465 on the client. If Unix extensions are enabled and the uids on the client
466 and server do not match, client permission checks are meaningless on
467 server uids that do not exist on the client (this does not affect the
468 normal ACL check which occurs on the server). Fix default uid
469 on mknod to match create and mkdir. Add optional mount parm to allow
470 override of the default uid behavior (in which the server sets the uid
471 and gid of newly created files). Normally for network filesystem mounts
472 user want the server to set the uid/gid on newly created files (rather than
473 using uid of the client processes you would in a local filesystem).
477 Make transaction counts more consistent. Merge /proc/fs/cifs/SimultaneousOps
478 info into /proc/fs/cifs/DebugData. Fix oops in rare oops in readdir
479 (in build_wildcard_path_from_dentry). Fix mknod to pass type field
480 (block/char/fifo) properly. Remove spurious mount warning log entry when
481 credentials passed as mount argument. Set major/minor device number in
482 inode for block and char devices when unix extensions enabled.
486 Fix /proc/fs/cifs/Stats and DebugData display to handle larger
487 amounts of return data. Properly limit requests to MAX_REQ (50
488 is the usual maximum active multiplex SMB/CIFS requests per server).
489 Do not kill cifsd (and thus hurt the other SMB session) when more than one
490 session to the same server (but with different userids) exists and one
491 of the two user's smb sessions is being removed while leaving the other.
492 Do not loop reconnecting in cifsd demultiplex thread when admin
493 kills the thread without going through unmount.
497 Do not rename hardlinked files (since that should be a noop). Flush
498 cached write behind data when reopening a file after session abend,
499 except when already in write. Grab per socket sem during reconnect
500 to avoid oops in sendmsg if overlapping with reconnect. Do not
501 reset cached inode file size on readdir for files open for write on
507 Update number of blocks in file so du command is happier (in Linux a fake
508 blocksize of 512 is required for calculating number of blocks in inode).
509 Fix prepare write of partial pages to read in data from server if possible.
510 Fix race on tcpStatus field between unmount and reconnection code, causing
511 cifsd process sometimes to hang around forever. Improve out of memory
512 checks in cifs_filldir
516 Fix incorrect file size in file handle based setattr on big endian hardware.
517 Fix oops in build_path_from_dentry when out of memory. Add checks for invalid
518 and closing file structs in writepage/partialpagewrite. Add statistics
519 for each mounted share (new menuconfig option). Fix endianness problem in
520 volume information displayed in /proc/fs/cifs/DebugData (only affects
521 affects big endian architectures). Prevent renames while constructing
522 path names for open, mkdir and rmdir.
526 Change to mempools for alloc smb request buffers and multiplex structs
527 to better handle low memory problems (and potential deadlocks).
531 Fix incomplete listings of large directories on Samba servers when Unix
532 extensions enabled. Fix oops when smb_buffer can not be allocated. Fix
533 rename deadlock when writing out dirty pages at same time.
537 Fix open of files in which O_CREATE can cause the mode to change in
538 some cases. Fix case in which retry of write overlaps file close.
539 Fix PPC64 build error. Reduce excessive stack usage in smb password
540 hashing. Fix overwrite of Linux user's view of file mode to Windows servers.
544 Fixes for large file copy, signal handling, socket retry, buffer
545 allocation and low memory situations.
549 Better port 139 support to Windows servers (RFC1001/RFC1002 Session_Initialize)
550 also now allowing support for specifying client netbiosname. NT4 support added.
554 Fix reconnection (and certain failed mounts) to properly wake up the
555 blocked users thread so it does not seem hung (in some cases was blocked
556 until the cifs receive timeout expired). Fix spurious error logging
557 to kernel log when application with open network files killed.
561 Fix /proc/fs module unload warning message (that could be logged
562 to the kernel log). Fix intermittent failure in connectathon
563 test7 (hardlink count not immediately refreshed in case in which
564 inode metadata can be incorrectly kept cached when time near zero)
568 Allow file_mode and dir_mode (specified at mount time) to be enforced
569 locally (the server already enforced its own ACLs too) for servers
570 that do not report the correct mode (do not support the
571 CIFS Unix Extensions).
575 Fix some small memory leaks in some unmount error paths. Fix major leak
576 of cache pages in readpages causing multiple read oriented stress
577 testcases (including fsx, and even large file copy) to fail over time.
581 Send NTCreateX with ATTR_POSIX if Linux/Unix extensions negotiated with server.
582 This allows files that differ only in case and improves performance of file
583 creation and file open to such servers. Fix semaphore conflict which causes
584 slow delete of open file to Samba (which unfortunately can cause an oplock
585 break to self while vfs_unlink held i_sem) which can hang for 20 seconds.
589 fixes to cifs_readpages for fsx test case
593 Fix caching data integrity bug when extending file size especially when no
594 oplock on file. Fix spurious logging of valid already parsed mount options
595 that are parsed outside of the cifs vfs such as nosuid.
600 Connect to server when port number override not specified, and tcp port
601 unitialized. Reset search to restart at correct file when kernel routine
602 filldir returns error during large directory searches (readdir).
606 Fix caching problem when files opened by multiple clients in which
607 page cache could contain stale data, and write through did
608 not occur often enough while file was still open when read ahead
609 (read oplock) not allowed. Treat "sep=" when first mount option
610 as an override of comma as the default separator between mount
615 Allow passwords longer than 16 bytes. Allow null password string.
619 Gracefully clean up failed mounts when attempting to mount to servers such as
620 Windows 98 that terminate tcp sessions during protocol negotiation. Handle
621 embedded commas in mount parsing of passwords.
625 Invalidate local inode cached pages on oplock break and when last file
626 instance is closed so that the client does not continue using stale local
627 copy rather than later modified server copy of file. Do not reconnect
628 when server drops the tcp session prematurely before negotiate
629 protocol response. Fix oops in reopen_file when dentry freed. Allow
630 the support for CIFS Unix Extensions to be disabled via proc interface.
634 Fix hang in commit_write during reconnection of open files under heavy load.
635 Fix unload_nls oops in a mount failure path. Serialize writes to same socket
636 which also fixes any possible races when cifs signatures are enabled in SMBs
637 being sent out of signature sequence number order.
641 Fix byte range locking bug (endian problem) causing bad offset and
646 Fix oops (in send_sig) caused by CIFS unmount code trying to
647 wake up the demultiplex thread after it had exited. Do not log
648 error on harmless oplock release of closed handle.
652 Fix unsafe global variable usage and password hash failure on gcc 3.3.1
653 Fix problem reconnecting secondary mounts to same server after session
654 failure. Fix invalid dentry - race in mkdir when directory gets created
655 by another client between the lookup and mkdir.
659 Fix to list processing in reopen_files. Fix reconnection when server hung
660 but tcpip session still alive. Set proper timeout on socket read.
664 Add missing mount options including iocharset. SMP fixes in write and open.
665 Fix errors in reconnecting after TCP session failure. Fix module unloading
666 of default nls codepage
670 Active smb transactions should never go negative (fix double FreeXid). Fix
671 list processing in file routines. Check return code on kmalloc in open.
672 Fix spinlock usage for SMP.
676 Fix oops in reopen_files when invalid dentry. drop dentry on server rename
677 and on revalidate errors. Fix cases where pid is now tgid. Fix return code
678 on create hard link when server does not support them.
682 Fix scheduling while atomic error in getting inode info on newly created file.
683 Fix truncate of existing files opened with O_CREAT but not O_TRUNC set.
687 Fix oops on write to dead tcp session. Remove error log write for case when file open
688 O_CREAT but not O_EXCL
692 Fix non-POSIX behavior on rename of open file and delete of open file by taking
693 advantage of trans2 SetFileInfo rename facility if available on target server.
694 Retry on ENOSPC and EAGAIN socket errors.
698 Fix oops on big endian readdir. Set blksize to be even power of two (2**blkbits) to fix
699 allocation size miscalculation. After oplock token lost do not read through
704 Fix oops on empty file readahead. Fix for file size handling for locally cached files.
708 Fix oops in mkdir when server fails to return inode info. Fix oops in reopen_files
709 during auto reconnection to server after server recovered from failure.
713 Finish support for Linux 2.5 open/create changes, which removes the
714 redundant NTCreate/QPathInfo/close that was sent during file create.
715 Enable oplock by default. Enable packet signing by default (needed to
716 access many recent Windows servers)
720 Fix oops when mounting to long server names caused by inverted parms to kmalloc.
721 Fix MultiuserMount (/proc/fs/cifs configuration setting) so that when enabled
722 we will choose a cifs user session (smb uid) that better matches the local
723 uid if a) the mount uid does not match the current uid and b) we have another
724 session to the same server (ip address) for a different mount which
725 matches the current local uid.
729 Add support for mknod of block or character devices. Fix oplock
730 code (distributed caching) to properly send response to oplock
735 Finish up CIFS packet digital signing for the default
736 NTLM security case. This should help Windows 2003
737 network interoperability since it is common for
738 packet signing to be required now. Fix statfs (stat -f)
739 which recently started returning errors due to
740 invalid value (-1 instead of 0) being set in the
741 struct kstatfs f_ffiles field.
745 Fix oops on stopping oplock thread when removing cifs when
750 Fix mount options for ro (readonly), uid, gid and file and directory mode.
754 Fix errors displayed on failed mounts to be more understandable.
755 Fixed various incorrect or misleading smb to posix error code mappings.
759 Fix display of NTFS DFS junctions to display as symlinks.
760 They are the network equivalent. Fix oops in
761 cifs_partialpagewrite caused by missing spinlock protection
762 of openfile linked list. Allow writebehind caching errors to
763 be returned to the application at file close.
767 Clean up options displayed in /proc/mounts by show_options to
768 be more consistent with other filesystems.
772 Fix delete of readonly file to Windows servers. Reflect
773 presence or absence of read only dos attribute in mode
774 bits for servers that do not support CIFS Unix extensions.
775 Fix shortened results on readdir of large directories to
776 servers supporting CIFS Unix extensions (caused by
777 incorrect resume key).
781 Fix truncate bug (set file size) that could cause hangs e.g. running fsx
785 unload nls if mount fails.
789 Add resume key support to search (readdir) code to workaround
790 Windows bug. Add /proc/fs/cifs/LookupCacheEnable which
791 allows disabling caching of attribute information for
796 Add more oplock handling (distributed caching code). Remove
797 dead code. Remove excessive stack space utilization from
802 Fix oops in get dfs referral (triggered when null path sent in to
803 mount). Add support for overriding rsize at mount time.
807 Fix buffer overrun in readdir which caused intermittent kernel oopses.
808 Fix writepage code to release kmap on write data. Allow "-ip=" new
809 mount option to be passed in on parameter distinct from the first part
810 (server name portion of) the UNC name. Allow override of the
811 tcp port of the target server via new mount option "-port="
815 Fix search handle leak on rewind. Fix setuid and gid so that they are
816 reflected in the local inode immediately. Cleanup of whitespace
817 to make 2.4 and 2.5 versions more consistent.
822 Fix signal sending so that captive thread (cifsd) exits on umount
823 (which was causing the warning in kmem_cache_free of the request buffers
824 at rmmod time). This had broken as a sideeffect of the recent global
825 kernel change to daemonize. Fix memory leak in readdir code which
826 showed up in "ls -R" (and applications that did search rewinding).
830 Reconnect tids and fids after session reconnection (still do not
831 reconnect byte range locks though). Fix problem caching
832 lookup information for directory inodes, improving performance,
833 especially in deep directory trees. Fix various build warnings.
837 Finish fixes to commit write for caching/readahead consistency. fsx
838 now works to Samba servers. Fix oops caused when readahead
839 was interrupted by a signal.
843 Fix data corruption (in partial page after truncate) that caused fsx to
844 fail to Windows servers. Cleaned up some extraneous error logging in
845 common error paths. Add generic sendfile support.
849 Fix memory leak in AllocMidQEntry.
850 Finish reconnection logic, so connection with server can be dropped
851 (or server rebooted) and the cifs client will reconnect.
855 Fix temporary socket leak when bad userid or password specified
856 (or other SMBSessSetup failure). Increase maximum buffer size to slightly
857 over 16K to allow negotiation of up to Samba and Windows server default read
858 sizes. Add support for readpages
862 Fix oops when username not passed in on mount. Extensive fixes and improvements
863 to error logging (strip redundant newlines, change debug macros to ensure newline
864 passed in and to be more consistent). Fix writepage wrong file handle problem,
865 a readonly file handle could be incorrectly used to attempt to write out
866 file updates through the page cache to multiply open files. This could cause
867 the iozone benchmark to fail on the fwrite test. Fix bug mounting two different
868 shares to the same Windows server when using different usernames
869 (doing this to Samba servers worked but Windows was rejecting it) - now it is
870 possible to use different userids when connecting to the same server from a
871 Linux client. Fix oops when treeDisconnect called during unmount on
872 previously freed socket.
876 Fix oops in readpages caused by not setting address space operations in inode in
881 Includes support for deleting of open files and renaming over existing files (per POSIX
882 requirement). Add readlink support for Windows junction points (directory symlinks).
886 Changed read and write to go through pagecache. Added additional address space operations.
887 Memory mapped operations now working.
891 Added writepage code for additional memory mapping support. Fixed leak in xids causing
892 the simultaneous operations counter (/proc/fs/cifs/SimultaneousOps) to increase on
893 every stat call. Additional formatting cleanup.
897 Fix bigendian bug in order of time conversion. Merge 2.5 to 2.4 version. Formatting cleanup.
901 Fixes from Zwane Mwaikambo for adding missing return code checking in a few places.
902 Also included a modified version of his fix to protect global list manipulation of
903 the smb session and tree connection and mid related global variables.
907 Fix problem with captive thread hanging around at unmount time. Adjust to 2.5.42-pre
908 changes to superblock layout. Remove wasteful allocation of smb buffers (now the send
909 buffer is reused for responses). Add more oplock handling. Additional minor cleanup.
913 More stylistic updates to better match kernel style. Add additional statistics
914 for filesystem which can be viewed via /proc/fs/cifs. Add more pieces of NTLMv2
915 and CIFS Packet Signing enablement.
919 Replace call to sleep_on with safer wait_on_event.
920 Make stylistic changes to better match kernel style recommendations.
921 Remove most typedef usage (except for the PDUs themselves).
925 Update mount so the -unc mount option is no longer required (the ip address can be specified
926 in a UNC style device name. Implementation of readpage/writepage started.
930 Fix intermittent problem with incorrect smb header checking on badly
931 fragmented tcp responses
935 Fixes to setting of allocation size and file size.
939 Various 2.5.38 fixes. Now works on 2.5.38
943 Prepare for 2.5 kernel merge. Remove ifdefs.
947 Socket buffer management fixes. Fix dual free.
951 Various big endian fixes for hardlinks and symlinks and also for dfs.
955 Various big endian fixes for servers with Unix extensions such as Samba
959 Various FindNext fixes for incorrect filenames on large directory searches on big endian
960 clients. basic posix file i/o tests now work on big endian machines, not just le
964 SessionSetup and NegotiateProtocol now work from Big Endian machines.
965 Various Big Endian fixes found during testing on the Linux on 390. Various fixes for compatibility with older
966 versions of 2.4 kernel (now builds and works again on kernels at least as early as 2.4.7).
970 Various minor fixes for Connectathon Posix "basic" file i/o test suite. Directory caching fixed so hardlinked
971 files now return the correct number of links on fstat as they are repeatedly linked and unlinked.
975 Implemented "Raw" (i.e. not encapsulated in SPNEGO) NTLMSSP (i.e. the Security Provider Interface used to negotiate
976 session advanced session authentication). Raw NTLMSSP is preferred by Windows 2000 Professional and Windows XP.
977 Began implementing support for SPNEGO encapsulation of NTLMSSP based session authentication blobs
978 (which is the mechanism preferred by Windows 2000 server in the absence of Kerberos).
982 Introduced optional mount helper utility mount.cifs and made coreq changes to cifs vfs to enable
983 it. Fixed a few bugs in the DFS code (e.g. bcc two bytes too short and incorrect uid in PDU).
987 Rewrote much of connection and mount/unmount logic to handle bugs with
988 multiple uses to same share, multiple users to same server etc.
992 Fixed major problem with dentry corruption (missing call to dput)
996 Rewrite of readdir code to fix bug. Various fixes for bigendian machines.
997 Begin adding oplock support. Multiusermount and oplockEnabled flags added to /proc/fs/cifs
998 although corresponding function not fully implemented in the vfs yet
1002 Fixed dentry caching bug, misc. cleanup
1006 Fixed 2.5 support to handle build and configure changes as well as misc. 2.5 changes. Now can build
1007 on current 2.5 beta version (2.5.24) of the Linux kernel as well as on 2.4 Linux kernels.
1008 Support for STATUS codes (newer 32 bit NT error codes) added. DFS support begun to be added.
1012 Unix extensions (symlink, readlink, hardlink, chmod and some chgrp and chown) implemented
1013 and tested against Samba 2.2.5
1018 1) Fixed lockrange to be correct (it was one byte too short)
1020 2) Fixed GETLK (i.e. the fcntl call to test a range of bytes in a file to see if locked) to correctly
1021 show range as locked when there is a conflict with an existing lock.
1023 3) default file perms are now 2767 (indicating support for mandatory locks) instead of 777 for directories
1024 in most cases. Eventually will offer optional ability to query server for the correct perms.
1026 3) Fixed eventual trap when mounting twice to different shares on the same server when the first succeeded
1027 but the second one was invalid and failed (the second one was incorrectly disconnecting the tcp and smb
1030 4) Fixed error logging of valid mount options
1032 5) Removed logging of password field.
1034 6) Moved negotiate, treeDisconnect and uloggoffX (only tConx and SessSetup remain in connect.c) to cifssmb.c
1035 and cleaned them up and made them more consistent with other cifs functions.
1037 7) Server support for Unix extensions is now fully detected and FindFirst is implemented both ways
1038 (with or without Unix extensions) but FindNext and QueryPathInfo with the Unix extensions are not completed,
1039 nor is the symlink support using the Unix extensions
1041 8) Started adding the readlink and follow_link code
git.openpandora.org / pandora-kernel.git / blob